ES & SD #'s & accuracy question **UPDATED 7-20-10**

[BigBamBoo]

Ok...I do not claim to be a professional re-loader. And I recently just bought a chronograph...a $100 ProChrono.

I have been working on some loads for a .243 and I get some pretty low ES and SD #'s (ES 13 and SD 5) but my best accuracy/groups come in with ES and SD # that are higher (ES 44 and SD 19).

I do my best to eliminate/limit any outside influences while doing load work. I shoot from a concrete bench, use a very stable front rest (JJ Machine), rear bag filled with heavy sand, and have my trigger set down to 6-8 oz. (Jewell). I keep my contact with the rifle to a minimum.

So...any thoughts or advice? I have always thought that the lower the ES and SD #'s the more consistent things should be and consistency = accuracy.

Take care,Stan

PS....one of my better five shot groups with the higher ES and SD #'s (shot at 100 yards)

 

[Bob3700]

Re: ES and SD #'s and accuracy question

Stan,

You are shooting at what most call point blank, short range. It seems that ES and SD are not that critical under 300 yds.

Obviously, your load and rifle are tuned for short range very well. Low ES and SD numbers, coupled with that same sort of accuracy is critical at longer ranges 600+ yds.

Low ES and SD numbers come from your loading techniques. Consistent neck tension, bullet seating depth, and a good combo of powder and primers.

If you can work on that load and get your numbers down, you will have a rifle capable of some excellent LR accuracy.

Keep up the good work.

Bob

 

[queen_stick]

Re: ES and SD #'s and accuracy question

read this...

http://www.accurateshooter.com/forum/index.php/topic,3746447.msg35857845.html#msg35857845

The posts by Rust and Laurie in that link are interesting, and well worth the time it takes to read them. There's a lot more going on with "ES vs. SD vs. Accuracy" than what's initially perceived.

Walt

 

[RonAKA]

Re: ES and SD #'s and accuracy question

At 100 yards a 44 fps difference will only move the poi about 0.07", so not a huge factor. What may be more significant is the actual velocity. For example if a "slow" bullet in the group just exits the barrel when it is at the maximum upward position, it will get the most "lift" from barrel vibration. A slightly faster bullet will exit the barrel sooner before the barrel tilt has peaked. This will move relative poi down some to correct for the extra velocity. This velocity range although there may be more shot to shot deviation in speed is somewhat self correcting.

On the other hand if a slow bullet exits the barrel just as it is at the maximum downward position, it will get an additional "drop" induced due to barrel position. A faster bullet will exit sooner and get less of a drop which coupled with more speed, it will be higher. This tuning magnifies the velocity variation effect.

So what you may want to try is taking the low ES load and increase or decrease velocity (powder) to see if that improves things.

A whole other subject, but it also takes a lot of samples to get an accurate standard deviation. Small samples can be misleading.

 

[Rust]

Re: ES and SD #'s and accuracy question

Inside 300 yards I wouldn't even bother with ES/SD if your rifle produces groups like that on a regular basis. For general target plinking or hunting use that's probably all the further you'll shoot.

When and if you start to stretch the range out it will matter. Then start with very accurately weighing your powder charges first (note: ONLY use powder from a single lot number for most accurate results), then trying different brands of primers (again from the same production lots for each primer), then different bullet seating depths (in effect changing the "compression ratio" but be careful there as it can cause pressures to rise seating shorter).

That would keep you busy for quite awhile...

 

[BigBamBoo]

Re: ES and SD #'s and accuracy question

Thank you all for the advice and help. This is for my 1000 yard rifle so I am trying to get things squared away for that.

Take care,Stan

 

[MarkS]

Re: ES and SD #'s and accuracy question

BigBamBoo,

Like Bob said. You would see your best load, would start to show vertical at long range. Shoot them at 200 or 300 yds and see what they look like. I like to see less than 20 ES for 600 and 1000 yds shooting.

Mark Schronce

 

[Bob3700]

Re: ES and SD #'s and accuracy question

OK,

This is for a 1K rifle. You are going to be shooting for ES under 30 FPS and SD in single digits.

Rust was right on with the single lot # of powder and experimenting with primers. CCI BR2, Wolf LR or SRM primers, Fed 210M and such are the one's that most have the best success with for 1K shooting.

Next, work on Neck Tension. Bullet release will be a big factor in lowering your numbers. I use very low neck tension and seat bullets into the lands. That way, the bullet is actually seated when the bolt closes and it should be the same every time.

Powder choice will be a factor. Use a powder that is know to produce good numbers in your cartridge. Varget, H4350, H4831 SC, VV 160, VV165 are some of the best for consistency.

ES and SD will be an indicator of your loading techniques for the most part. YOu will work and you will learn.

Bob

 

[Rust]

Re: ES and SD #'s and accuracy question

Speaking of neck tension...

Uniform neck tension from round to round is very important as it contributes to a uniform powder burn.

Just starting out, get your brass from a single production lot number and then keep it segragated by number of firings, the brass work hardens and neck tension increases for the same amount of resizing. At some point the brass will need to be annealed to maintain being able to maintain proper tension.

I have found that neck tension is the last piece of the puzzle for attaining very low ES/SD numbers once everything else has been optimized.

That'll keep you busy the rest of the year....

 

[bozo699]

Re: ES and SD #'s and accuracy question

I Know how to figure avg fps and extreme spread, but how do you figure sd if you know the velocity string?
Wayne.

 

[camac]

Re: ES and SD #'s and accuracy question

To confuse you and calculate SD you first work out the average,
then for each velocity in the string you square the difference from the average,
then add all these together,
then divide by number of shots in string and
finally take the square root.

e.g. 3010, 3015, 3020, 3025, 3030.
Mean (average) is 3020
Differences from mean are -10, -5, 0, 5, 10
Squares of these are 100, 25, 0, 25, 100
Sum of these squares are 250.
Divide by number of shots (5) is 50
Square root of 50 is 7.07 (SD) and ES of 20

 

[RonAKA]

Re: ES and SD #'s and accuracy question

Another way if you have Excel is to use the STDEV function. What is important is what it means.

In this graph future velocities of the same load would be expected to fall within the red area if you take the average plus and minus the standard deviation. This is about 68%. If you take the average plus and minus three standard deviations, then 99% of future velocities will be within that range (all the way out to the end of the blue). So you have some confidence future velocities will be within 1 standard deviation, and really good confidence they will be within 3 standard deviations.

 

[GermanS1]

Re: ES and SD #'s and accuracy question

For those who are generally interested in the topic of statistics for shooters, this is one one the best articles I've read on the topic. I'm very thankful to the author Jerry Engelman for his permission to post it on my site.

http://riflemansjournal.blogspot.com/2010/07/statistics-for-rifle-shooters.html

 

[camac]

Re: ES and SD #'s and accuracy question

German, Great article by Jerry. I would love to see someone apply this logic to the target as well. In general we judge a rifle on group sizes. In reality we should give them a s.d. and ES as well (thus basically explaining the distirbution curve of the load/rifle). The age old argument of 3 shot vs 5 shot vs 10 shot groups is explained really well by statistics and normal distribution curves like RonAKA posted above. It is just a whole lot harder to measure and calculate in the field. Even with the widest bell curve, statistically speaking there is a reasonable chance you could still shoot a tight three shot group. In reality we should collect data on a particular load from all shots (good and bad) and put them into the data set to give a more "accurate" assessment of how the rifle and load is performing.
Software like on target would only need an added formula to calculate this for us.

To give an example of this, Would you throw away a load that had a goup size of 0.5moa but 9/10 shots were within 0.2. This type of target is common and if you scored purely on group size with no notes you might throw it away - or look back later on spreadsheet and gloss over it. A lot of shooters will note "fliers" but they happen regularly for nearly all of us. This target would have a very low SD but higher ES.

Interestingly as a score shooter I am more interested in SD (or how close most of my shots are grouped) rather than ES (the flier) but am forever chasing the reasons for the fliers. World records are broken when both are achieved (Like Jackie Schmidts amazing result).
Cam

 

[GermanS1]

Re: ES and SD #'s and accuracy question

Cam, you're correct in your analysis and there are two good ways to count all shots on paper. The first is simply score and we do it all the time. I look at a load based on X count more than anything. If I can get 12 X's from 20 shots at 500 or 600 yards, then I consider the load a good performer. The stray 9's are my fault, missed wind call, poor execution, etc.

The other way is to measure mean radius and that's how Frankford Arsenal and Lake City did it for decades. They calculated mean radius for twenty seven 10 shot groups for each lot of ammo. When I do it I only measure mean vertical dispersion, with the wind we have here, I ignore horizontal.

To measure mean vertical dispersion, draw an arbitrary horizontal line across the target a few inches below the lowest shot. Measure and record the vertical distance from the line to each shot. Calculate the mean and draw a new horizontal line at that point. Using the original measurements, subtract the distance from the first line to the new line from each measurement, ignore negative signs, and calculate the mean from the true centerline. It's easier than it sounds. That's how I do all my primer testing on paper.

 

[camac]

Re: ES and SD #'s and accuracy question

Nice idea German. As you said it is easier to find vertical centre than finding the overall centre of a group. Although "on target does it somehow?...
We also shoot long range for score in quite variable conditions. Only thing is a lot of wind comes straight down our range and a 20-30 knot (or worse) breeze switching from 1 oclock to 11 oclock (and eddying) can have a fair bit of effect vertically as well. But it wouldn't be smart to develop loads in these conditions anyway - just makes the competition interesting.

Sorry for hijacking original context but interesting nonetheless.

 

[Steve Blair]

Re: ES and SD #'s and accuracy question

The folks at statshooting.com have an array of tools and papers explaining shot dispersion analysis. It may be more than most shooters want but is similar to techniques used at arsenals for ammunition analysis.

 

[camac]

Re: ES and SD #'s and accuracy question

Thanks Steve, This is the sort of stuff I was looking for. A bit heavy for this old brain but will try and dig into it. Cheers.

 

[JHardy]

Re: ES and SD #'s and accuracy question

BBB:

I hope this will help you a little. Take a good read of the article linked by German. It is a good one. Long range accuracy is about removing the vertical. If the vertical is gone (for the most part) it is all about reading the conditions.

The short range group is a good one. But, here is a way to analyze the group further. Take a good look. The windage is pretty good -- which means you had a good bag setup and table technique. Howver, the group has verticle in it. It could be that you need a little more powder; it could be that you need a little more neck tension; you may need to tune the seating depth; your forend may be not be far enough out past the front bag which can cause some flex and vertical; the front bag may be packed too tight with sand (dry fire the rifle on the bags and see if the x-hairs jump); your rifle may perform better with different shouldler pressure or free recoil; etc.

Don't let a small group in and of itself fool you. Look at the shape. You want a tune that gives you a round group. You are very close.

An example of ES and SD: Assume for a minute that you are in a class of 20 students. The class average is a 93, but your average is an 85. A visiting professor comes in the class and picks you to represent this fine class against another professor's class in a short academic competition. The other 19 students would probably like another class member to be the representative -- this is SD -- good bottom line -- not so good on one bad shot.

At llong range score shooting, you would rather shoot two 9s than one 8. You would rather be in a class where all 20 stucents had a 93 average and were equal representatives. The average is the same, but the outcome could be a lot different if the 20th shot (the poor stucent) is counted on to make the X ring. If you have a great ES, your SD will always be great. A great SD does not guarantee a great ES on EACH shot. I always figured I could read SD on the print out sheet and ES on the long range target. If you are average, you might be as close to the bottom as you are to the top. Think about it.

Just food for thought and another way to look at SD -- through the back door looking at ES.

Jim Hardy

 

[camac]

Re: ES and SD #'s and accuracy question

Jim, trying to work out whether BBB means Bull__t, Bull__t, Bull__t or Blood Brain Barrier or Boring Beyond Belief - can never get these right and found a heap more to confuse me. http://www.all-acronyms.com/BBB (32 of them).

Assuming Boring Beyond Belief and to try and avoid the other meanings as much as possible.

I think the main point of the article is that 3 or 5 shot strings through a chrony can give you an indication but you need much larger sample sizes to work out ES and SD correctly. You can actually have a low ES in a 5 shot string from a load that isn't actually doing that well and subsequently be fooled by the results. Looking at Ronakas nice colourful diagram on the previous page. A high percentage of shots fall within a very close space to the average. This means there is a good chance on any 3- 5 shot string that all shots can fall within that range. The difference in the accuracy to which a sample size can get a good result is amazing. 3 shots are very week, 5 get better, 10 better still but statistically speaking you start to get a good indication around 20-30+ shots. By then you have a good chance of seeing your "fliers" or 5% of the population or "extreme spread".

Group sizes can be considered exactly the same.

The problem for us shooters is we don't want to burn out barrels doing the testing.